3results about How to "Accurate interpolation" patented technology

The application provides a speech enhancement method and device based on a hidden space Schrodinger bridge, equipment and a medium. The method comprises the following steps: obtaining degraded speech; encoding the degraded speech into degraded latent variables in a hidden space through a target encoder, the target encoder being used to converge different types of degraded speech to a distribution close to corresponding clean latent variables in the hidden space, and the target encoder having the function of maintaining the energy characteristics of the power spectrum of an audio signal; processing the degraded latent variables through a target generation network based on a Schrodinger bridge to obtain clean latent variables corresponding to the degraded latent variables; and decoding the clean latent variables through a target decoder to obtain clean speech corresponding to the degraded speech. The application can solve the problem that the related art cannot realize the structured reconstruction of missing spectral information in a high-noise and low signal-to-noise ratio environment, and effectively improve the speech enhancement effect.

The invention discloses a swirler downstream fire core trajectory prediction method and system, electronic equipment and a medium, and the method comprises the steps: carrying out the flame time sequence state deduction of any two adjacent flame images through a flame time sequence state deduction model, and carrying out the prediction of a flame image of an intermediate time sequence; the method comprises the following steps: acquiring a flame image sequence, correspondingly interpolating the flame image sequence into an original flame image sequence to reconstruct and obtain a new flame image sequence, and generating a fire core track based on the new flame image sequence, considering that the establishment of a fire core depends on the distribution of pixel gray values in a spatial dimension, and performing feature learning and time sequence state deduction on the flame image sequence to obtain a flame core track. The method achieves the reconstruction of a flame image at an intermediate moment, achieves the more precise interpolation of a fire core track, achieves the precise prediction of a continuous track of a fire core in a time sequence, improves the positioning precision of the fire core while reducing the high-frame-rate sampling demands, reduces the dependence on high-cost imaging hardware, and improves the prediction accuracy of the fire core track.

This disclosure provides an adaptive interpolation method and electronic device for carbon emission power density at the city block level, relating to the field of data processing technology. The implementation scheme is as follows: Based on the differences in the gradient vectors of building development volume at various sampling points of a first urban plot, the location of a first interpolation sampling point in the sampling network of the first urban plot is determined, where the building development volume of the sampling point is the ratio of the building area to the base area of ​​the sampling point; based on the carbon emission power density of multiple first neighboring sampling points and the slope of the change in carbon emission power density of the multiple first neighboring sampling points, the carbon emission power density of the first interpolation sampling point is determined; based on the location and carbon emission power density of the first interpolation sampling point, interpolation is performed on the sampling network of the first urban plot. Using the scheme of this disclosure, the accuracy of carbon emission data interpolation can be improved, and the sampling cost of carbon emission data can be reduced.